Postinterventional passive expansion of partially dilated transjugular intrahepatic portosystemic shunt stents

Abstract

Purpose: To investigate passive transjugular intrahepatic portosystemic shunt (TIPS) stent expansion in patients with intentional "underdilation" (eg, 10-mm stent, 8-mm balloon) during TIPS creation.

Materials and methods: Custom in-house software was developed for objective quantification of cross-sectional stent area from computed tomography (CT) data. The technique was validated by in vitro experiments. The study included 39 patients (22 men; mean age, 59.2 y) who underwent TIPS creation (VIATORR stent graft [W. L. Gore & Associates, Flagstaff, Arizona]; n = 29; WALLSTENT endoprosthesis [Boston Scientific, Marlborough, Massachusetts], n = 10) with stent underdilation. Follow-up CT data of the patients were used to quantify in vivo stent area changes. Data were analyzed by variance analysis and entered into a general linear model to test for interrelations between stent area changes and clinical (eg, cirrhosis grade) and procedural parameters.

Results: In vitro validation of the in-house software showed good agreement and reproducibility without overestimation of stent area. Mean clinical follow-up time in patients was 787 days (range, 7-2,450 d). At the time of intervention, VIATORR stent grafts and WALLSTENT endoprostheses were dilated to an average of 64.4% ± 2.3% and 65.63% ± 8.52% of nominal area, respectively. At the last imaging follow-up evaluation, this value had increased in all stents to a mean of 87.8% ± 7.9% (VIATORR) and 82.34% ± 19.6% (WALLSTENT) in the TIPS tract (P < .05). Multivariate analysis revealed the time after intervention to be the only predictor of stent area in the TIPS tract. There was no significant association between stent expansion and clinical or procedure-related parameters.

Conclusions: The area of self-expanding stents implanted in the liver for TIPS creation with dilation to less than nominal diameter significantly increases over time. This increase has to be considered as an additional factor influencing the long-term portosystemic gradient.